Change of thermal conductivity of polyurethane pre

Transcription

Change of thermal conductivity of polyurethane pre
Published in Euroheat & Power Fernwärme International 6/99, June 1999
Change of thermal conductivity of polyurethane
pre-insulated pipes as a function of time
JÜRGEN KELLNER, VEERLE DIRCKX
Huntsman Polyurethanes
Everslaan 45
B-3078 Everberg
Belgium
Tel.: +32 (0)2 758 9420
Summary
The increase of thermal conductivity was determined on pre-insulated district heating pipe composites
which were exposed to an accelerated ageing. Two types of rigid polyurethane foams were used: a fully
water blown and a cyclopentane blown foam system. The experimental results show that the thermal
conductivity increase depends on the pipe ageing conditions. The results can be useful to check various
models for the prediction of the thermal conductivity increase of pre-insulated pipes as a function of time.
Zusammenfassung des Berichts:
Veränderung der Wärmeleitfähigkeit als Funktion der Zeit
und verschiedener Alterungsbedingungen bei Polyurethan
vorgedämmten Rohren
Eine Vielzahl von Gründen sprechen für den Einsatz von
Polyurethan-Hartschäumen bei vorisolierten KunststoffMantelrohren im Fernwärmebereich, z.B. die ausgezeichnete
mechanische Festigkeit, die hohe thermische Belastbarkeit,
die geringe Wasseraufnahme sowie die vorzügliche
Dämmeigenschaft. Allerdings verschlechtern sich die
Dämmeigenschaften im Laufe der Zeit durch
Gasdiffusionsprozesse und dadurch kann die Wirtschaftlichkeit
eines Fernwärmenetzes beeinträchtigt werden. Daher ist es
von allgemeinem Interesse, die Grössenordnung der Abnahme
der Dämmeigenschaften, in anderen Worten, die Zunahme
der Wärmeleitfähigkeit von Polyurethan-Schäumen in
Fernwärmerohren zu untersuchen.
In einem Forschungsprogramm wurde die Zunahme der
Wärmeleitfähigkeit von vorgedämmten DN50 KunststoffMantelrohren bestimmt, die unter drei verschiedenen
Bedingungen gealtert wurden. Dabei wurden zwei
Schaumtypen untersucht: ein wasser-getriebenes System
sowie ein cyclopentan-getriebenes System. Der Einfluss
des Kunststoffmantels als Diffusionssperre wurde ebenfalls
untersucht, indem die Änderung der Wärmeleitfähigkeit an
Rohren mit und ohne HDPE-Mantelrohr gemessen wurde.
Die Zunahme der Wärmeleitfähigkeit wurde bei folgenden
Bedingungen ermittelt: homogene Temperaturbelastung von
50°C Innen- und Aussentemperatur sowie 70°C Innen-und
Aussentemperatur, und inhomogene Temperaturbelastung
von 175°C Innen- und 25°C Aussentemperatur. Folgende
Schlussfolgerungen können gezogen werden:
– das HDPE Mantelrohr ist eine sehr wirksame
Diffusionbarriere für Kohlendioxid, das durch Reaktion
aus Isozyanat mit Wasser entsteht. Die Zunahme
der Wärmeleitfähigkeit ist bei wasser-getriebenen
Systemen in Gegenwart eines ca. 3 mm dicken
HDPE-Mantels erheblich verzögert.
– bei cyclopentan-getriebenen Schaumsystemen
ist das HDPE-Mantelrohr für den Erhalt der
Dämmeigenschaften weit weniger wichtig.
Im Vergleich zu Kohlendioxid und Luft diffundiert
Cyclopentan sehr langsam durch den Schaum,
so dass der Schaum selbst als Diffusionsbarriere
angesehen werden kann.
page 2
– die anfängliche Wärmeleitfähigkeit eines wassergetriebenen Schaumes erhöht sich um ca. 10 mW/m.K
aufgrund von Gasdiffusion, d.h. Ersatz von
Kohlendioxid durch Luft, das als Zellgas schlechtere
Dämmeigenschaften aufweist als Kohlendioxid.
Bei cyclopentan-getriebenen Systemen erhöht sich
die anfängliche Wärmeleitfähigkeit um 4-5 mW/m.K.
– je höher die Alterungs-Temperatur, desto schneller
erfolgt der Anstieg der Wärmeleitfähigkeit.
– Gasdiffusionsprozesse in vorgedämmten Rohren
laufen rascher ab bei einer homogenen Alterung,
d.h. wenn Innen- und Aussentemperaturbelastung
gleich sind, als bei einer inhomogenen Alterung,
d.h. wenn die Aussentemperatur ist deutlich
niedriger ist als die Rohr-Innentemperatur.
Im Vergleich zu der beschriebenen forcierten Alterung im
Experiment ist die Gasdiffusion in der Praxis bei erdverlegten
Rohren erheblich reduziert. Folglich ist auch die Zunahme
der Wärmeleitfähigkeit deutlich langsamer.
– Erde selbst fungiert als Dämmmaterial und reduziert
die Gasdiffusion. Ausserdem ist die Konzentration
von Luft, die in die Zellen diffundieren kann, in der
Erde deutlich geringer.
– in diesem Forschungsprojekt wurden DN50-Rohre
verwendet. Bei einem grösseren Rohrdurchmesser
kann man davon ausgehen, dass die Diffusionsvorgänge
verlangsamt ablaufen, da sowohl das Mantelrohr
als auch die Dämmschicht dicker sind.
– da es sich hier um eine forcierte Alterung handelt, sind
die im Experiment verwendeten Alterungstemperaturen
deutlich höher als die Bedingungen in der Praxis.
Bei der Abnahme der Langzeitdämmfähigkeit von erdverlegten
Fernwärmerohren handelt es sich um einen sehr langsamen
Prozess. Die hier beschriebenen Ergebnisse können von Nutzen
sein, verschiedene mathematische Modelle zu testen, um eine
durch Diffusionsprozesse bedingte Verschlechterung der
Dämmeigenschaften von Polyurethan-Hartschäumen in
Kunststoff-Mantelrohren abzuschätzen.
Introduction
There are a number of reasons why rigid polyurethane foam is used as
insulation material in the district heating area, e.g. the high mechanical
strength, the high thermal resistance, the ability to fill narrow cavities,
the low water absorption, the good adhesion to service and casing pipe
and foremost, the excellent thermal insulation property.
The excellent insulation will efficiently minimise the heat loss during the transport of hot water or steam in pre-insulated pipes in
a district heating network. The insulation performance however will deteriorate during the service life of a pipe, due to gas diffusion
processes leading to changes in the cell gas composition. This is referred to as thermal conductivity ageing and results in an increased
energy loss during transport of hot water which may reduce the profitability of a district heating network. Therefore, an assessment
of the increase of the thermal conductivity as a function of time is of paramount importance for pipe producers and district
heating network owners.
We have carried out thermal conductivity measurements on polyurethane pre-insulated DN50 pipes which were aged at three
different temperatures. Two types of foam were evaluated, a fully water blown foam system and a water/cyclopentane dual blown
system. The ageing was done on pipes with and without HDPE casing to evaluate the effect of the casing pipe as a diffusion barrier.
The results obtained are described in the following. We believe that they are useful to validate mathematical models for the
prediction of the thermal conductivity change of pre-insulated pipes over time.
Why is rigid polyurethane foam a good insulating material
Rigid polyurethane foam is a good insulator because it consists of 92 to 98% of closed cells which are filled
with insulating gases. Only 8 to 2% of the foam is solid polyurethane polymer. The percentage of solid polymer
is determined by the density of the foam: the higher the density of the foam, the higher the percentage of solid
polymer. The closed cells are filled with several gases released by blowing agents during the manufacture
of the polyurethane foam.
The cell gas composition can be influenced by using different blowing agents. This has a major impact on the thermal conductivity of
the foam, as the composition of the gas determines more than 60% of the final thermal conductivity of the foam at the applied density.
One can lower the thermal conductivity of the foam by choosing blowing agents that have lower vapour thermal conductivity, see also
Table 1 which compares typical foam systems with different blowing agents used for pipe insulation. Table 2 lists the thermal
conductivity of various gases [1].
Rigid Polyurethane foam is used in pipe insulation since it provides
excellent mechanical strength
combined with
excellent thermal insulation properties
page 3
Thermal conductivity ageing of polyurethane foam
A fresh foam is characterised by a certain insulation value: the initial thermal conductivity value. According
to the EN 253 norm for polyurethane pre-insulated pipes, the thermal conductivity should be maximum
33.0 mW/m.K at 50°C [2]. However, this initial value will not be maintained over time because gases are
able to migrate into or out of the foam. This causes changes in the cell gas composition and hence in the
insulation value of the foam. The result is a gradually increasing thermal conductivity value of the foam.
Each gas has its own diffusion rate through polyurethane foam:
– In the short term, carbon dioxide will diffuse out
of the foam and will be replaced by air. An effect
on the thermal conductivity can be seen because
air is characterised by a higher thermal conductivity
compared to carbon dioxide. A net increase of the
thermal conductivity of the foam system will occur.
– In the long term, the physical blowing agent will
again escape out of the cells. However, in general
the physical blowing agents are considered to be
permanent gases and have very low diffusion speeds.
Due to their long term presence in the cells, though
the concentration will drop, the thermal conductivity
of foams expanded by physical blowing agents
will increase at a much slower rate.
The gas diffusion phenomena occurs until the gases partial
pressures inside and outside the foam are equal. The processes
are accelerated by higher temperatures and are decelerated
when making use of diffusion barriers that are impermeable
or semi-permeable for gases.
There is limited experimental data available concerning gas
diffusion processes in district heating pipe composites and
thus the thermal conductivity ageing of district heating pipes.
λ=
φ ln (D2 /D0)
2πL(T0 –T2)
This prompted us to initiate a research project where pipe
samples were subjected to accelerated ageing. Pipe segments
with fully water blown and water/cyclopentane dual blown
foams were homogeneously aged in ovens at 50°C and 70°C,
and in-homogeneously on a pipe rack with an outside
temperature of 25°C and an inside temperature of 175°C,
i.e. by circulation of steam in the service pipe. To quantify
the effect of the HDPE casing, the ageing was carried out
with and without casing and the thermal conductivity change
was recorded as a function of time. The density of the
polyurethane foam was ca. 85 kg/cm3.
The thermal conductivity of DN50 pipe composites was
measured at an average temperature of 50°C with a so-called
Kapapipe, designed and made by A.L.B. Instrumentation,
St.-Ismier, France. One metre long DN50 pre-insulated pipe
segments were used to record the thermal flux in the middle
35cm of the pipe, after a stable temperature gradient between
inner and outer pipe was established, i.e. 80°C inside and 20°C
outside. This temperature gradient causes a heat flow from the
higher temperature to the lower temperature. This heat flow is
proportional to the insulating performance of the pipe as well
as to the magnitude of the temperature gradient. The thermal
conductivity of the pipe can be calculated if the energy needed
to maintain the inner steel pipe at the higher temperature and
the temperature of inner and outer pipe are known:
with: – φ the power needed to maintain the temperature of the steel pipe
– D2 and D0 respectively the inner and outer diameter of the pipe
– T2 and T0 respectively the temperatures of inner and outer pipe
– L the length of the pipe segment
The equipment was calibrated by means of a DN50 pipe filled with expanded polystyrene, for which the thermal conductivity value was
determined at different temperatures by the Department of Building Physics at the University of Chalmers, Gothenburg in Sweden.
page 4
Experimental results and conclusions
The results of our thermal conductivity measurements on polyurethane pre-insulated pipes subjected to
homogeneous ageing at 50°C and 70°C and in-homogeneous ageing at 25°C outside temperature and 175°C
inside temperature are shown in Tables 3-5 and Figures 1-3. The following general conclusions can be drawn:
– the HDPE casing is a very efficient diffusion barrier
for air and carbon dioxide: the thermal conductivity
ageing rate of water blown pipes stored at 70°C is
significantly lower when a 3mm thick HDPE casing
is present. The HDPE casing is therefore crucial for
maintaining the insulating performance of a pipe
with a water blown foam.
– the HDPE casing is less important for the cyclopentane
blown foams: the physical blowing agent migrates
extremely slowly through the foam, compared to
air and carbon dioxide. Hence, the foam itself can
be regarded as a diffusion barrier. The thermal
conductivity ageing rate is only marginally lower
when a 3mm thick casing is applied.
– the initial thermal conductivity value of a water blown
pipe will increase by approximately 10 mW/m.K due
to diffusion. For the physical blown pipes this increase
is 4 to 5 mW/m.K.
– the higher the temperature, the faster the thermal
conductivity ageing.
– the diffusion processes in pipes exposed to
in-homogeneous ageing are slower compared to the
pipes which were homogeneously aged in an oven.
Our experimental results are similar to accelerated pipe
ageing results obtained at DTI, Denmark [3]. However,
it should also be noted that these measurements, both from
DTI and ourselves, reflect a higher thermal conductivity
ageing rate than is encountered in reality:
– the fact that polyurethane pre-insulated pipes are
buried in the ground considerably reduces the speed of
the diffusion processes as the soil acts as an insulation
layer but also as the presence of air is limited.
– for the research project DN50 pipes were used.
For bigger diameter pipes, the diffusion will be
decelerated as there is usually a thicker HDPE
casing and a thicker foam layer present.
– the ageing conditions as applied in the research
project are much more severe than in reality,
i.e. a higher temperature gradient was used.
This is also confirmed by measurements carried out on a
pre-insulated pipe with a water blown foam which was buried
in the ground. The temperature of the transported water
was 50-60°C during summer and 80-90°C during winter.
The thermal conductivity value after 8 years service
increased only slightly by 2.3 mW/m.K, from 31.6 mW/m.K
to 33.9 mW/m.K which is an increase of 7% [4].
In reality, the thermal conductivity ageing is a very slow
process which takes years. Therefore it is interesting to predict
the ageing profiles with computer models. We believe that this
experimental data for the thermal conductivity change as a
function of ageing temperature and time can be of value to
assess various mathematical models for the prediction of the
thermal conductivity change over time in pre-insulated pipes.
page 5
HDPE is a very efficient
diffusion barrier
for air and carbon dioxide
page 6
Blowing agent
Average initial
thermal conductivity of DN50
pipe composites at 50°C (mW/m.K)
Water blown
31
Cyclopentane/water dual blown
28
HCFC-141b/water dual blown
28
Gas
Gas thermal conductivity
at 25°C (mW/m.K)
Air
25.9
Carbon dioxide
16.2
Cyclopentane
13.0
HCFC-141b
10.0
Table 1: Typical thermal conductivity of
pipe insulation foam systems
Tabelle 1: Typische Wärmeleitfähigkeit von
Hartschaumsystemen eingesetzt im
Fernwärmebereich
Table 2: Thermal conductivity of typical
cell gases at 25°C
Tabelle 2: Wärmeleitfähigkeit typischer
Zellgase bei 25°C
Ageing time
at 50°C
(days)
water blown,
no HDPE casing
(mW/m.K)
water blown
(mW/m.K)
c-pentane blown,
no HDPE casing
(mW/m.K)
c-pentane blown
(mW/m.K)
0
29.3
29.1
28.0
27.1
40
31.0
29.8
29.5
27.8
83
32.2
30.3
29.6
28.1
114
33.7
31.1
30.7
29.0
151
36.3
31.8
30.1
28.1
201
37.1
32.5
31.4
28.4
270
39.4
33.9
–
29.3
Ageing time
at 70°C
(days)
water blown,
no HDPE casing
(mW/m.K)
water blown
(mW/m.K)
c-pentane blown,
no HDPE casing
(mW/m.K)
c-pentane blown
(mW/m.K)
0
30.0
30.8
29.0
28.2
1
30.4
–
–
–
9
31.3
–
–
–
15
33.1
–
29.0
–
22
34.2
–
–
–
30
35.4
31.4
–
29.1
37
–
–
30.1
–
54
37.9
–
30.6
–
60
38.3
32.1
–
28.7
68
–
–
30.2
–
70
38.5
–
–
–
84
39.1
–
–
–
94
–
–
31.7
–
110
40.2
–
–
–
150
–
35.5
–
29.9
154
–
–
32.3
–
170
40.4
–
–
–
210
–
36.4
33.4
29.7
260
41.8
38.4
33.7
31.1
297
–
37.3
34.6
31.3
330
–
37.2
33.2
31.4
350
40.5
–
–
–
365
–
38.2
33.7
31.2
421
–
38.4
33.2
31.8
455
41.8
–
–
–
526
–
39.9
39.9
32.8
Table 3: λ-increase for DN50
pipe composites homogeneously
aged at 50°C
Tabelle 3: λ-Zunahme bei
DN50-Rohren gealtert bei 50°C
Innen- und Aussentemperatur
Table 4: λ-increase for DN50
pipe composites homogeneously
aged at 70°C
Tabelle 4: λ-Zunahme bei
DN50-Rohren gealtert bei 70°C
Innen- und Aussentemperatur
page 7
The -ageing rate of cyclopentane
blown foams
is considerably slower compared to
water blown foams
Table 5: λ-increase for
DN50 pipe composites
homogeneously aged at
an outside temperature
of 25°C and inside
temperature of 175°C
Tabelle 5: λ-Zunahme bei
DN50-Rohren gealtert bei
25°C Aussen- und 175°C
Innentemperatur
page 8
Ageing time
at 25°C/175°C
(days)
water blown,
no HDPE casing
(mW/m.K)
water blown
(mW/m.K)
c-pentane blown,
no HDPE casing
(mW/m.K)
c-pentane blown
(mW/m.K)
0
29.7
29.6
28.2
27.7
35
37.2
30.7
29.1
29.0
64
–
–
30.0
–
67
38.4
31.5
–
29.0
97
40.3
31.4
–
29.5
114
–
–
30.4
–
147
39.9
33.0
–
29.7
200
–
–
33.0
–
230
40.6
37.7
–
31.0
305
–
–
34.7
–
335
40.9
39.0
–
32.3
Figure 1
Figure 1: λ-increase for DN50 pipe
composites homogeneously aged at 50°C
Bild 1: λ-Zunahme bei DN50-Rohren
gealtert bei 50°C Innen- und
Aussentemperatur
Figure 2
Figure 2: λ-increase for DN50 pipe
composites homogeneously aged at 70°C
Bild 2: λ-Zunahme bei DN50-Rohren
gealtert bei 70°C Innen- und
Aussentemperatur
Figure 3
Figure 3: λ-increase for DN50 pipe
composites homogeneously aged at
an outside temperature of 25°C and
inside temperature of 175°C
Bild 3: λ-Zunahme bei DN50-Rohren
gealtert bei 25°C Aussen- und 175°C
Innentemperatur
page 9
page 10
References
Biographies
1. M. Svanström, “Blowing agents in rigid polyurethane foam”,
Doctoral Thesis 1997, Department of Chemical
Environmental Science at Chalmers University of
Technology, Gothenburg, Sweden
2. European Standard EN 253, “Pre-insulated bonded pipe
systems for underground hot water networks – pipe
assembly of steel service pipes, polyurethane thermal
insulation and outer casing of polyethylene”, 1994
3. H. Smidt, J. Daugaard, “Long-term insulating properties
of pre-insulated district heating pipes”, Euroheat & Power,
4-5/1997, 140-146
4. U. Jarfelt, “Field measurements of gas diffusion from district
heating pipes”, P-98:6, Nov. 1998, Department of Building
Technology at Chalmers University of Technology,
Gothenburg, Sweden
Jürgen Kellner
Dr. Jürgen Kellner, Shell Research and Technology Centre,
Louvain-la-Neuve in Belgium, responsible for the applicational
research, development activities and technical service in
rigid polyurethane foam insulated pipes. The author joined
Huntsman Polyurethanes in 1999 following a strategic alliance
in the area of rigid polyurethane between Shell Chemicals
and Huntsman Polyurethanes.
Veerle Dirckx
Dr. Veerle Dirckx, until January 1999 employed by Shell
Research and now working for Artilat, Nijlen in Belgium,
responsible for research and development in flexible
urethanes and latex foams.
The information, technical data and recommendations in this paper are, to the best of our knowledge, reliable. Tests performed and
referred to in the paper do not necessarily represent all possible uses or actual performance as this is very much dependent on the
particular circumstances the product or foam is used in. Suggestions made concerning the products and their uses, applications,
storage and handling are only the opinion of the Huntsman Polyurethanes group and users should make their own tests to determine
the suitability of these products for their own particular purpose. Huntsman Polyurethanes makes no guarantee or warranty of any
kind, whether express or implied, other than that the product conforms to its applicable Standard Specifications. Statements made
herein, therefore, should not be construed as representations or warranties.
‘Daltofoam’ and ‘Suprasec’ are trademarks of Huntman ICI Chemicals LLC.